Abstract
The therapeutic alliance is considered a robust predictor of psychotherapy outcome. Ruptures and resolutions in the alliance have been the focus of recent alliance literature. Most previous studies investigated their between-patient effects. We used hierarchical linear models to disaggregate the between- and within-patient effects of ruptures on the alliance from patient- and therapist perspective and symptom severity. Further, the moderating effects of rupture resolutions were analyzed. The sample consisted of 56 patients diagnosed with depressive or anxiety disorders who received 25 ± 3 sessions of integrative cognitive behavioral therapy in the outpatient setting. The observer-rated rupture resolution rating system was applied to all 1st, 8th, 16th and 24th therapy sessions. The alliance quality and symptom severity were assessed using self-report questionnaires after these four sessions. Results showed that a higher intensity of ruptures across treatment was associated with lower alliance ratings from both patient- and therapist perspectives during treatment. After sessions with more intense confrontation ruptures, both reported a weaker alliance. Rupture resolutions significantly moderated the withdrawal rupture effect on the alliance. The results provide meaningful practical implications for therapist feedback and training. They further underline the importance of using appropriate statistical analyses to the data structure and nature of psychotherapy to better understand the role of the alliance, rupture, and repair during therapy.